{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:LIRRLUPPLJQ7OFJDY3PUMPVUHM","short_pith_number":"pith:LIRRLUPP","schema_version":"1.0","canonical_sha256":"5a2315d1ef5a61f71523c6df463eb43b0b253d2c2cd7278c93da12c344414579","source":{"kind":"arxiv","id":"1804.10969","version":3},"attestation_state":"computed","paper":{"title":"UNIQ: Uniform Noise Injection for Non-Uniform Quantization of Neural Networks","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alex M. Bronstein, Avi Mendelson, Chaim Baskin, Eli Schwartz, Evgenii Zheltonozhskii, Natan Liss, Raja Giryes","submitted_at":"2018-04-29T17:38:20Z","abstract_excerpt":"We present a novel method for neural network quantization that emulates a non-uniform $k$-quantile quantizer, which adapts to the distribution of the quantized parameters. Our approach provides a novel alternative to the existing uniform quantization techniques for neural networks. We suggest to compare the results as a function of the bit-operations (BOPS) performed, assuming a look-up table availability for the non-uniform case. In this setup, we show the advantages of our strategy in the low computational budget regime. While the proposed solution is harder to implement in hardware, we beli"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"1804.10969","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2018-04-29T17:38:20Z","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"title_canon_sha256":"d8335f7220dffcdbb03475a332060c7ab0585ba154726908a4fb0ac76f03f404","abstract_canon_sha256":"933aca8926c49cf74b41f6f41e835bdd84806af8511d6f4245842fbb5cc5387c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:26:44.664024Z","signature_b64":"ZCQr2l02gJEOD6P0jwSZ6T2ROfJAys9M22mGSmTE7swsd9lzT9Sz6mGmhjpg5rBQy3be8dRNxh5l24CbyhVDDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5a2315d1ef5a61f71523c6df463eb43b0b253d2c2cd7278c93da12c344414579","last_reissued_at":"2026-07-05T02:26:44.663474Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:26:44.663474Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"UNIQ: Uniform Noise Injection for Non-Uniform Quantization of Neural Networks","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alex M. Bronstein, Avi Mendelson, Chaim Baskin, Eli Schwartz, Evgenii Zheltonozhskii, Natan Liss, Raja Giryes","submitted_at":"2018-04-29T17:38:20Z","abstract_excerpt":"We present a novel method for neural network quantization that emulates a non-uniform $k$-quantile quantizer, which adapts to the distribution of the quantized parameters. Our approach provides a novel alternative to the existing uniform quantization techniques for neural networks. We suggest to compare the results as a function of the bit-operations (BOPS) performed, assuming a look-up table availability for the non-uniform case. In this setup, we show the advantages of our strategy in the low computational budget regime. While the proposed solution is harder to implement in hardware, we beli"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1804.10969","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/1804.10969/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"1804.10969","created_at":"2026-07-05T02:26:44.663532+00:00"},{"alias_kind":"arxiv_version","alias_value":"1804.10969v3","created_at":"2026-07-05T02:26:44.663532+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1804.10969","created_at":"2026-07-05T02:26:44.663532+00:00"},{"alias_kind":"pith_short_12","alias_value":"LIRRLUPPLJQ7","created_at":"2026-07-05T02:26:44.663532+00:00"},{"alias_kind":"pith_short_16","alias_value":"LIRRLUPPLJQ7OFJD","created_at":"2026-07-05T02:26:44.663532+00:00"},{"alias_kind":"pith_short_8","alias_value":"LIRRLUPP","created_at":"2026-07-05T02:26:44.663532+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1908.05033","citing_title":"Differentiable Soft Quantization: Bridging Full-Precision and Low-Bit Neural Networks","ref_index":3,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LIRRLUPPLJQ7OFJDY3PUMPVUHM","json":"https://pith.science/pith/LIRRLUPPLJQ7OFJDY3PUMPVUHM.json","graph_json":"https://pith.science/api/pith-number/LIRRLUPPLJQ7OFJDY3PUMPVUHM/graph.json","events_json":"https://pith.science/api/pith-number/LIRRLUPPLJQ7OFJDY3PUMPVUHM/events.json","paper":"https://pith.science/paper/LIRRLUPP"},"agent_actions":{"view_html":"https://pith.science/pith/LIRRLUPPLJQ7OFJDY3PUMPVUHM","download_json":"https://pith.science/pith/LIRRLUPPLJQ7OFJDY3PUMPVUHM.json","view_paper":"https://pith.science/paper/LIRRLUPP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1804.10969&json=true","fetch_graph":"https://pith.science/api/pith-number/LIRRLUPPLJQ7OFJDY3PUMPVUHM/graph.json","fetch_events":"https://pith.science/api/pith-number/LIRRLUPPLJQ7OFJDY3PUMPVUHM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LIRRLUPPLJQ7OFJDY3PUMPVUHM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LIRRLUPPLJQ7OFJDY3PUMPVUHM/action/storage_attestation","attest_author":"https://pith.science/pith/LIRRLUPPLJQ7OFJDY3PUMPVUHM/action/author_attestation","sign_citation":"https://pith.science/pith/LIRRLUPPLJQ7OFJDY3PUMPVUHM/action/citation_signature","submit_replication":"https://pith.science/pith/LIRRLUPPLJQ7OFJDY3PUMPVUHM/action/replication_record"}},"created_at":"2026-07-05T02:26:44.663532+00:00","updated_at":"2026-07-05T02:26:44.663532+00:00"}